Please see below for details of a one-day workshop (lectures plus practical sessions) on GWAS methods by University College London Genetics Institute.
For more information or to register, visit this link
————————
Description
Date: Tuesday 23rd October, University College London, 10am-4.30pm
Tutors: Prof David Balding (Melbourne and UGI) and Dr Doug Speed (Aarhus and UGI)
Cost: £40 (or £30 for UCL Members)
Advance Registration is REQUIRED
——————————
Background: In recent years there has been great progress in developing genome-wide statistical analyses for detecting causal variants, constructing prediction models and better understanding the genetic architecture of complex traits. However the underlying regression models involve very large numbers of predictors, and strong modelling assumptions are required to tackle the consequent problem of over-fitting. The results can be sensitive to these assumptions, and also to the effects of population structure, genotyping errors and the extent to which rare SNPs are included. Recently-developed methods based on summary statistics are susceptible to similar problems
Course outline: We will cover genome-wide association analysis, including latest developments in confounding adjustments, and heritability analyses, both using individual-level genetic data (GCTA, LDAK) and using summary statistics (LDSC, SumHer). We will also cover assessing heritability enrichment in functionally-annotated regions, genetic correlation and risk prediction (e.g., polygenic risk scores, BLUP and MultiBLUP). We will emphasise the common elements of these methods, highlighting a standard framework that has emerged for genome-wide SNP analysis, while also contrasting the differences in modelling assumptions underlying the different software.
The practicals will provide step-by-step details for analysing genetic data, starting either with individual-level data (e.g., PLINK files or the output from IMPUTE2) or summary statistics (p-values from a GWAS). There will be a selection of worked examples; to take part in the practicals, participants should bring a laptop with either MAC or LINUX OS
Prerequisites: Participants should be proficient in statistics including some familiarity with random-effects regression models. In genetics, knowledge of SNP genotypes and Hardy-Weinberg and linkage equilibrium will be assumed. Computer scripts and output will be discussed that assume some familiarity with scientific computing using linux. Some familiarity with PLINK would be helpful but is not essential.
——————————
Provisional Timetable
10:00 – 12:20: Lecture 1 followed by Practical 1
Introduction to analysing GWAS data analysis using individual genotype data, kinship and heritability, both classical and SNP-based. Effect of LD, MAF and genotyping quality on heritability. GCTA and LDAK software.
12:20 – 13:00: Lunch
13.00 – 14:40: Lecture 2 followed by Practical 2
Methods based on summary statistics, assessing the effects of confounding in association analysis, enrichment of functional categories. LDSC, SumHer softwares
14:40 – 15:00: Break
15.00 – 16:30: Lecture 3 followed by Practical 3
Genetic correlations, genomic prediction and enhanced polygenic risk scores.